CN113139925A - Pneumonia image processing method, system and storage medium - Google Patents
Pneumonia image processing method, system and storage medium Download PDFInfo
- Publication number
- CN113139925A CN113139925A CN202110448471.9A CN202110448471A CN113139925A CN 113139925 A CN113139925 A CN 113139925A CN 202110448471 A CN202110448471 A CN 202110448471A CN 113139925 A CN113139925 A CN 113139925A
- Authority
- CN
- China
- Prior art keywords
- image
- pneumonia
- processed
- enhanced
- reconstructed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 206010035664 Pneumonia Diseases 0.000 title claims abstract description 97
- 238000003672 processing method Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000001914 filtration Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 19
- 230000004927 fusion Effects 0.000 claims description 25
- 239000011159 matrix material Substances 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 abstract description 5
- 238000012795 verification Methods 0.000 abstract description 3
- 238000012549 training Methods 0.000 description 11
- 238000013527 convolutional neural network Methods 0.000 description 9
- 238000012360 testing method Methods 0.000 description 7
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 210000004072 lung Anatomy 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000011976 chest X-ray Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20208—High dynamic range [HDR] image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30061—Lung
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a pneumonia image processing method, a pneumonia image processing system and a storage medium, wherein a pneumonia image to be processed is subjected to filtering reconstruction and feature enhancement and then is fused with an original pneumonia image, so that the characteristics of the original pneumonia image and the image subjected to feature enhancement are reserved, and subsequent neural network learning is facilitated. Through verification of the Inception V3 network, the pneumonia image is processed by the method, and the obtained pneumonia image is improved in accuracy and specificity compared with an unprocessed pneumonia image and a pneumonia image processed by only using a Retinex algorithm.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a pneumonia image processing method, a pneumonia image processing system, and a storage medium.
Background
In the case of new coronary pneumonia, researchers in this field in China have proposed different methods for rapid identification of pneumox images. However, since the deep learning training is carried out by depending on the features extracted by the convolutional neural network, a plurality of scholars aim at the optimization of the convolutional neural network. The preprocessing of the X-ray image still remains in simple operations such as denoising and enhancing.
Although the X-ray images obtained by the traditional methods of denoising, filtering, histogram equalization and the like are feature-enhanced in human vision and can be used for judging diseases more easily, the difference between the normal lung images and the diseased lung images is not obvious, and the original fine features are lost due to the traditional preprocessing method and are not beneficial to neural network learning.
Disclosure of Invention
The embodiment of the invention provides a pneumonia image processing method, a pneumonia image processing system and a storage medium, which are used for solving the problems that in the prior art, a traditional preprocessing method causes fine feature loss and is not beneficial to neural network learning.
In one aspect, an embodiment of the present invention provides a pneumonia image processing method, including:
filtering and reconstructing a pneumonia image to be processed to obtain a reconstructed image;
performing feature enhancement on the reconstructed image to obtain an enhanced image;
and performing characteristic fusion on the pneumonia image to be processed and the enhanced image to obtain a fusion image.
In a possible implementation manner, performing filtering reconstruction on the pneumonia image to be processed to obtain a reconstructed image may include: constructing an edge centralization matrix; and performing convolution processing on the pneumonia image to be processed by using an edge centering matrix to obtain a reconstructed image.
In one possible implementation, the edge centering matrix may be a third order matrix with a center value of 0.00 and edge data of 0.125.
In one possible implementation, the performing feature enhancement on the reconstructed image to obtain an enhanced image may include: and (5) performing characteristic enhancement on the reconstruction appearance by adopting a single-scale Retinex algorithm to obtain an enhanced image.
In a possible implementation manner, performing feature enhancement on the reconstructed image by using a single-scale Retinex algorithm to obtain an enhanced image may include: decomposing the reconstructed image into an incident image and a reflected image; and reducing the influence of the incident image on the reconstructed image, obtaining the reflection attribute of the reconstructed image and obtaining the enhanced image.
In one possible implementation manner, performing feature fusion on the pneumonia image to be processed and the enhanced image to obtain a fused image may include: setting the weight ratio of the pneumonia image to be processed and the enhanced image to be 0.50, and setting the threshold value to be 0.00; and performing feature fusion on the pneumonia image to be processed and the enhanced image by adopting the set weight proportion and the set threshold value to obtain a fusion image.
In another aspect, an embodiment of the present invention provides a pneumonia image processing system, including:
the reconstruction module is used for carrying out filtering reconstruction on the pneumonia image to be processed to obtain a reconstructed image;
the enhancement module is used for carrying out feature enhancement on the reconstructed image to obtain an enhanced image;
and the fusion module is used for performing characteristic fusion on the pneumonia image to be processed and the enhanced image to obtain a fusion image.
In another aspect, an embodiment of the present invention provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores computer instructions executable by the at least one processor to cause the at least one processor to perform the method described above.
In another aspect, an embodiment of the present invention provides a computer-readable storage medium, in which a plurality of computer instructions are stored, and the computer instructions are used to enable a computer to execute the above method.
In another aspect, an embodiment of the present invention provides a computer program product, and when being executed by a processor, the computer program product implements the method described above.
The pneumonia image processing method, the pneumonia image processing system and the storage medium have the following advantages:
after filtering reconstruction and characteristic enhancement are carried out on the pneumonia image to be processed, the pneumonia image to be processed is fused with the original pneumonia image, so that the characteristics of the original pneumonia image and the image after characteristic enhancement are reserved, and subsequent neural network learning is facilitated.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of a pneumonia image processing method according to a first embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating an image processing effect of the pneumonia image processing method according to the present invention;
fig. 3 is a flowchart of a pneumonia image processing method according to a second embodiment of the present invention;
fig. 4 is a flowchart of a pneumonia image processing method according to a third embodiment of the present invention;
fig. 5 is a flowchart of a pneumonia image processing method according to a fourth embodiment of the present invention;
fig. 6 is a flowchart of a pneumonia image processing method according to a fifth embodiment of the present invention;
fig. 7 is a functional block diagram of a pneumonia image processing system according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the prior art, researchers in China put forward different methods for rapidly identifying pneumonia X-ray images, and most of the methods focus on a convolutional neural network. Before the convolutional neural network is used for image learning, the image needs to be preprocessed, and the conventional preprocessing method is simple, so that some details are lost in the processing process, and the learning accuracy of the convolutional neural network is reduced.
Aiming at the problems in the prior art, the invention provides a pneumonia image processing method, a pneumonia image processing system and a storage medium, wherein a pneumonia image to be processed is subjected to filtering reconstruction and characteristic enhancement and then is fused with an original pneumonia image, so that the characteristics of the original pneumonia image and the image with the enhanced characteristics are reserved, and the subsequent neural network learning is facilitated. Through verification of an Inception V3 network, the pneumonia image is processed by the method, and the accuracy and specificity of the obtained pneumonia image are improved compared with an unprocessed pneumonia image and a pneumonia image processed by only using a Retinex algorithm.
Fig. 1 is a flowchart of a pneumonia image processing method according to a first embodiment of the present invention. The pneumonia image processing method provided by the embodiment of the invention comprises the following steps:
and S100, filtering and reconstructing the pneumonia image to be processed to obtain a reconstructed image.
Illustratively, the pneumonia image to be processed is an X-ray image. The purpose of filtering and reconstructing the pneumonia image is to retain the detail characteristics of the pneumonia image and simultaneously suppress the noise in the pneumonia image. Common filter reconstruction methods include: linear filtering and nonlinear filtering, the linear filtering mainly comprises mean filtering and gaussian filtering, and the nonlinear filtering mainly comprises median filtering and bilateral filtering. The mean filtering is to replace the value of a certain pixel in the original image with the mean value of all pixels in a region, and the filtering reconstruction method adopted in the embodiment of the invention is the mean filtering method.
And S101, performing feature enhancement on the reconstructed image to obtain an enhanced image.
Illustratively, there are two goals for feature enhancement of images: the method has the advantages that firstly, the visual effect of the image is improved, the definition of the image is improved, secondly, the interested characteristics are highlighted aiming at the application occasion of the given image, the uninteresting characteristics are restrained, the difference between the characteristics of different objects in the image is enlarged, and some special analysis requirements are met. The current image feature enhancement method comprises the following steps: spatial domain based methods and frequency domain based methods. The spatial domain-based method is to directly process the pixels of the image, and the frequency domain-based method is to modify the transform coefficient of the image in a certain transform domain of the image and then inversely transform the image to the original spatial domain to obtain an enhanced image.
And S102, performing feature fusion on the pneumonia image to be processed and the enhanced image to obtain a fusion image.
Illustratively, after two times of processing in S100 and S101, the obtained enhanced image inevitably loses some details, which help the learning of the convolutional neural network, so that the learning of the convolutional neural network is adversely affected by using the enhanced image alone. The enhanced image and the pneumonia image to be processed are subjected to feature fusion, so that the detail features in the original pneumonia image and the detail features of the enhanced image can be reserved, and convenience is brought to subsequent convolutional neural network learning, as shown in fig. 2.
After the method is adopted, the pneumonia image processing method provided by the embodiment of the invention has the following beneficial effects:
1. compared with the method that X-rays are acquired by professionals to judge pneumonia, more time is saved and efficiency is improved by the method from registration to X-ray image acquisition.
2. Compared with the classification of the deep learning images by unprocessed images, the method has the advantages that the precision is more accurate, and the misdiagnosis probability is reduced to a greater extent.
Fig. 3 is a flowchart of a pneumonia image processing method according to a second embodiment of the present invention. In a possible embodiment, S100, performing filtering reconstruction on the pneumonia image to be processed to obtain a reconstructed image, may include: s300, constructing an edge centralization matrix; s301, performing convolution processing on the pneumonia image to be processed by using the edge centering matrix to obtain a reconstructed image.
Illustratively, the edge centering matrix is a third order matrix with a center value of 0.00 and edge data of 0.125, as follows:
after the edge centering matrix is adopted to carry out convolution processing on the pneumonia image to be processed, the average value of 8 pixels at the edge of a certain pixel in the pneumonia image to be processed is used for replacing the value of the pixel.
Fig. 4 is a flowchart of a pneumonia image processing method according to a third embodiment of the present invention. In a possible embodiment, S101, performing feature enhancement on the reconstructed image to obtain an enhanced image may include: and S400, performing feature enhancement on the reconstructed image by adopting a single-scale Retinex algorithm to obtain an enhanced image.
Illustratively, Retinex is a commonly used image enhancement method, which was proposed by edwin. The basic theory of Retinex theory is: the color of the object is determined by the reflection capacity of the object to long-wave (red), medium-wave (green) and short-wave (blue) light rays, but not by the absolute value of the intensity of the reflected light, and the color of the object is not influenced by illumination nonuniformity and has uniformity, namely Retinex is based on color sense uniformity. Unlike the traditional linear and nonlinear methods which can only enhance a certain feature of an image, Retinex can balance three aspects of dynamic range compression, edge enhancement and color constancy, so that various different types of images can be adaptively enhanced. For more than 40 years, researchers imitate the human visual system to develop a Retinex algorithm, and the Retinex algorithm is improved from a single-scale Retinex algorithm to a multi-scale weighted average (MSR) algorithm, and then the color recovery multi-scale MSRCR algorithm is developed.
Fig. 5 is a flowchart of a pneumonia image processing method according to a fourth embodiment of the present invention. In a possible embodiment, S400, performing feature enhancement on the reconstructed image by using a single-scale Retinex algorithm to obtain an enhanced image, which may include: s500, decomposing the reconstructed image into an incident image and a reflected image; s501, reducing the influence of the incident image on the reconstructed image, obtaining the reflection attribute of the reconstructed image, and obtaining the enhanced image.
Illustratively, a Single Scale Retinex (SSR) algorithm is used in the embodiments of the present invention, which is based on the theory that the image seen by human eyes is the reflection of light on an object. Assuming one image is S (x, y), it can be decomposed into two different images: reflection image R (x, y) and incident image L (x, y), also called luminance image:
S(x,y)＝R(x,y)*L(x,y)
the theory of the Retinex algorithm is to obtain the reflection properties of objects in an image by reducing the influence of an incident image L (x, y) on an original image S (x, y), and further enhance the image. The formula from which the SSR algorithm can be derived is:
r(x,y)＝logS(x,y)-logL(x,y)
taking L (x, y) ═ F (x, y) × S (x, y), where
Is the center surround function, c is the gaussian surround scale, and λ is an automatically derived scale. The value of λ satisfies ═ F (x, y) dxdy ═ 1. And taking c as 300, and processing the image.
Fig. 6 is a flowchart of a pneumonia image processing method according to a fifth embodiment of the present invention. In a possible embodiment, S102, performing feature fusion on the pneumonia image to be processed and the enhanced image to obtain a fused image, which may include: s600, setting the weight proportion of the pneumonia image to be processed and the weight proportion of the enhancement image to be 0.50, and setting the threshold value to be 0.00; s601, performing feature fusion on the pneumonia image and the enhanced image to be processed by adopting the set weight proportion and the set threshold value to obtain a fusion image.
By way of example, after the parameter setting is adopted, features in the pneumonia image to be processed and the enhanced image can be reserved to the maximum extent, and therefore the fused image can be used as convolutional neural network learning.
Performance verification
In order to prove that the method provided by the invention is really effective, experiments are carried out below to verify.
The experiments were performed under the Windows 10 professional edition:
a processor: intel (R) core (TM) i5-8500 CPU @3.00GHz
Memory: 8.00GB
The system type is as follows: 64 bit operating system, x64 based processor
A frame: tensorflow framework of Windows edition
Before the start of the experiment, a Chest-X-Ray Image pneumonia Image data set published by Kermany et al in 2018 was collected, including common lung images and pneumonia images. Dividing the images in the data set into a training set and a testing set according to a certain proportion, wherein the number of the images in the two sets is respectively as follows:
ChestX-Ray Image dataset distribution
At the beginning of the experiment, Google open source inclusion V3 is used as a test network, images in a training set are input into the test network, two 3 x 3 networks are used in the network to replace a 5 x 5 network, one 1 x 3 network and one 3 x 1 network are used to replace a 3 x 3 network, and under the condition that the receptive field is kept unchanged, the network nonlinearity is greatly enhanced. The network structure is as follows:
incep V3 network architecture
In the training process, in order to prevent the images in the training set from randomly influencing the training result and ensure the stability of the training result, 3 times of training are carried out on the back 20-layer network of the inclusion V3 network at the same time under the same CPU environment, and the average value of the 3 times of training is used as the final training result. After the network training is completed, the data in the test set are respectively processed by the following three processes: keeping unchanged, only adopting a single-scale Retinex algorithm to carry out image enhancement, adopting the method provided by the invention to carry out processing, respectively inputting the images obtained by the three processing methods into a trained test network, and outputting results of three aspects of accuracy, specificity and sensitivity:
average value of Incep V3 network test results
In the three indexes, the sensitivity is the accurate ratio predicted in the pneumonia cases, and the specificity is the accurate ratio predicted in the non-pneumonia cases. From the above test results, it can be seen that the method of the present invention has a great improvement in accuracy and specificity, and a small reduction in sensitivity, compared to the original image and the image processed only by the SSR algorithm. It is true that the method of the invention can produce certain advantages.
The invention also provides a pneumonia image processing system, which comprises:
the reconstruction module 700 is configured to perform filtering reconstruction on the pneumonia image to be processed to obtain a reconstructed image;
an enhancement module 701, configured to perform feature enhancement on the reconstructed image to obtain an enhanced image;
and the fusion module 702 is configured to perform feature fusion on the pneumonia image to be processed and the enhanced image to obtain a fusion image.
The present invention also provides an electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores computer instructions executable by the at least one processor to cause the at least one processor to perform the method described above.
The present invention also provides a computer-readable storage medium having stored thereon a plurality of computer instructions for causing a computer to perform the above-described method.
The invention also provides a computer program product, which when executed by a processor implements the method described above.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A pneumonia image processing method characterized by comprising:
filtering and reconstructing a pneumonia image to be processed to obtain a reconstructed image;
performing feature enhancement on the reconstructed image to obtain an enhanced image;
and performing characteristic fusion on the pneumonia image to be processed and the enhanced image to obtain a fusion image.
2. The pneumonia image processing method according to claim 1, wherein the filtering reconstruction of the pneumonia image to be processed to obtain a reconstructed image comprises:
constructing an edge centralization matrix;
and performing convolution processing on the pneumonia image to be processed by using the edge centering matrix to obtain the reconstructed image.
3. The pneumonia image processing method according to claim 2, wherein the edge centering matrix is a third-order matrix with a central value of 0.00 and edge data of 0.125.
4. The pneumonia image processing method according to claim 1, wherein the performing feature enhancement on the reconstructed image to obtain an enhanced image includes:
and performing characteristic enhancement on the reconstruction appearance by adopting a single-scale Retinex algorithm to obtain the enhanced image.
5. The pneumonia image processing method according to claim 4, wherein the performing feature enhancement on the reconstructed image by using a single-scale Retinex algorithm to obtain the enhanced image comprises:
decomposing the reconstructed image into an incident image and a reflected image;
and reducing the influence of the incident image on the reconstructed image, obtaining the reflection attribute of the reconstructed image, and obtaining the enhanced image.
6. The pneumonia image processing method according to claim 1, wherein the feature fusion of the pneumonia image to be processed and the enhanced image to obtain a fusion image comprises:
setting the weight ratio of the pneumonia image to be processed and the enhanced image to be 0.50, and setting a threshold value to be 0.00;
and performing feature fusion on the pneumonia image to be processed and the enhanced image by adopting the set weight proportion and the threshold value to obtain the fused image.
7. A pneumonia image processing system characterized by comprising:
the reconstruction module is used for carrying out filtering reconstruction on the pneumonia image to be processed to obtain a reconstructed image;
the enhancement module is used for carrying out characteristic enhancement on the reconstructed image to obtain an enhanced image;
and the fusion module is used for performing characteristic fusion on the pneumonia image to be processed and the enhanced image to obtain a fusion image.
8. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein
The memory stores computer instructions executable by the at least one processor to cause the at least one processor to perform the method of any one of claims 1-6.
9. A computer-readable storage medium having stored thereon a plurality of computer instructions for causing a computer to perform the method of any one of claims 1-6.
10. A computer program product, characterized in that the computer program realizes the method of any of claims 1-6 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110448471.9A CN113139925B (en) | 2021-04-25 | 2021-04-25 | Pneumonia image processing method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110448471.9A CN113139925B (en) | 2021-04-25 | 2021-04-25 | Pneumonia image processing method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113139925A true CN113139925A (en) | 2021-07-20 |
| CN113139925B CN113139925B (en) | 2024-04-12 |
Family
ID=76811915
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110448471.9A Active CN113139925B (en) | 2021-04-25 | 2021-04-25 | Pneumonia image processing method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113139925B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110188639A (en) * | 2019-05-20 | 2019-08-30 | 深圳供电局有限公司 | Face image processing method and system, computer equipment and readable storage medium |
| EP3579180A1 (en) * | 2018-06-07 | 2019-12-11 | Beijing Kuangshi Technology Co., Ltd. | Image processing method and apparatus, electronic device and non-transitory computer-readable recording medium for selective image enhancement |
| CN110706174A (en) * | 2019-09-27 | 2020-01-17 | 集美大学 | Image enhancement method, terminal equipment and storage medium |
| CN110969622A (en) * | 2020-02-28 | 2020-04-07 | 南京安科医疗科技有限公司 | Image processing method and system for assisting pneumonia diagnosis |
| WO2021043273A1 (en) * | 2019-09-04 | 2021-03-11 | 华为技术有限公司 | Image enhancement method and apparatus |
-
2021
- 2021-04-25 CN CN202110448471.9A patent/CN113139925B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3579180A1 (en) * | 2018-06-07 | 2019-12-11 | Beijing Kuangshi Technology Co., Ltd. | Image processing method and apparatus, electronic device and non-transitory computer-readable recording medium for selective image enhancement |
| CN110188639A (en) * | 2019-05-20 | 2019-08-30 | 深圳供电局有限公司 | Face image processing method and system, computer equipment and readable storage medium |
| WO2021043273A1 (en) * | 2019-09-04 | 2021-03-11 | 华为技术有限公司 | Image enhancement method and apparatus |
| CN110706174A (en) * | 2019-09-27 | 2020-01-17 | 集美大学 | Image enhancement method, terminal equipment and storage medium |
| CN110969622A (en) * | 2020-02-28 | 2020-04-07 | 南京安科医疗科技有限公司 | Image processing method and system for assisting pneumonia diagnosis |
Non-Patent Citations (2)
| Title |
|---|
| 卢玮;高涛;王翠翠;陈本豪;张赛;: "Retinex理论下基于融合思想的低照度彩色图像增强算法", 科学技术与工程, no. 13 * |
| 袁甜;程红阳;陈云虹;张海荣;王文军;: "基于U-Net网络的肺部CT图像分割算法", 自动化与仪器仪表, no. 06 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113139925B (en) | 2024-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110399929B (en) | Fundus image classification method, fundus image classification apparatus, and computer-readable storage medium | |
| Zhang et al. | Single image defogging based on multi-channel convolutional MSRCR | |
| CN105654436B (en) | A kind of backlight image enhancing denoising method based on prospect background separation | |
| Wang et al. | Low illumination color image enhancement based on Gabor filtering and Retinex theory | |
| CN108830818B (en) | Rapid multi-focus image fusion method | |
| CN112381897B (en) | Low-illumination image enhancement method based on self-coding network structure | |
| CN113837974B (en) | NSST domain power equipment infrared image enhancement method based on improved BEEPS filtering algorithm | |
| Chen et al. | Remote sensing image quality evaluation based on deep support value learning networks | |
| Wang et al. | Low-light image joint enhancement optimization algorithm based on frame accumulation and multi-scale Retinex | |
| CN111415304A (en) | Underwater vision enhancement method and device based on cascade deep network | |
| CN111210395A (en) | Retinex underwater image enhancement method based on gray value mapping | |
| CN104318529A (en) | Method for processing low-illumination images shot in severe environment | |
| Nnolim | Smoothing and enhancement algorithms for underwater images based on partial differential equations | |
| CN116757986A (en) | Infrared and visible light image fusion method and device | |
| CN117611501A (en) | Low-illumination image enhancement method, device, equipment and readable storage medium | |
| CN115797205A (en) | Unsupervised single image enhancement method and system based on Retinex fractional order variation network | |
| Gui et al. | Adaptive single image dehazing method based on support vector machine | |
| CN116645305A (en) | Low-light image enhancement method based on multi-attention mechanism and Retinex | |
| CN116091357A (en) | Low-light image enhancement method for fusion of depth convolution attention and multi-scale features | |
| Yang et al. | CSDM: A cross-scale decomposition method for low-light image enhancement | |
| Tan et al. | Image-dehazing method based on the fusion coding of contours and colors | |
| Zhou et al. | An improved algorithm using weighted guided coefficient and union self‐adaptive image enhancement for single image haze removal | |
| CN111445422B (en) | Random asymptotic photon mapping image noise reduction method and system based on neural network | |
| Xi et al. | Enhancement of unmanned aerial vehicle image with shadow removal based on optimized retinex algorithm | |
| CN114066786A (en) | Infrared and visible light image fusion method based on sparsity and filter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |